Astron. Astrophys. 320, L37-L40 (1997)

3. Results

Even though GRS1915+105 was observed in the pointed-mode for 4 days during the period July 20-29, 1996 the useful data are for about 8850 seconds due to various operational constraints. Almost all the count rates were acquired with an integration time of 0.1 sec. The X-ray light curve for the entire period of useful observations showed no large scale intensity variations on time scale of a minute or longer. This is unlike the results reported in the higher energy band (Castro-Tirado et al. 1994 ; Sazonov et al. 1994 ). Analysis of data on shorter time scale, however, shows pronounced variations on time scale of a second and less. A few typical light curves showing sub-second intensity variations are shown in figure 1. These variations were detected independently in each detector with similar count rate profiles. Further the veto layer count rates do not exhibit this kind of variability. It will be noticed from the light curves that GRS 1915+105 shows frequent flaring activity on time scales of less than a second and occasionally over 0.1 sec. During the flares the intensity varies by a factor of upto 3 in less than a second. From analysis of the flare frequency in GRS1915+105 and Cyg X-1, we find that flare occurrence in GRS1915+105 is less common than that in Cyg X-1. Details of flaring activity in Cyg X-1 will be reported in a separate paper (Rao et al. 1997). During the sporadic intensity variations, the light curves of the two sources exhibit remarkable similarity at time scale of 0.1 to 1 second. Power density spectrum and Fast Fourier Transform (FFT) analysis showed no variability with a fixed period indicating that the source is not an X-ray pulsar. However, strong Quasi- Periodic Oscillation (QPOs) were detected from GRS1915+105 in all the data. The periodogram is shown in figure 2. The QPOs are clearly detected independently at the same frequency in the data of each PPC as well as in the summed data. The QPO frequency however varies in an erratic manner from day to day. A summary of the observations and QPO characteristics is given in Table I. It will be noticed that the QPO frequency varied from 0.62 to 0.82 Hz. The rms fraction in QPO is typically about 10%. Besides the 0.7 Hz peak the summed power spectrum also shows another less prominent peak at 1.4 Hz, which can be the first harmonic of the main peak. The power spectrum is nearly flat between 0.02 to 0.5 Hz with a power law index of -0.26 and becomes steep above the QPO frequency with a power law index of -1.26 in the 2 to 5 Hz range.

Fig. 1. The subsecond flux variations seen in GRS1915+105 with the Indian X-ray Astronomy Experiment (IXAE). Observations were made with 100 msec time resolution during 24-27 July 1996. Similar bursts were seen in all the detectors and the combined data are plotted here. Each panel shows a few of the flares seen on the days mentioned in the figure. Data of 3 seconds are plotted with a bin size of 100 msec around the flares. A factor of 2 or more increase in the X-ray flux is seen for a duration of about 100 to 400 msec.

Fig. 2. The Power density spectrum of the X-ray source GRS1915+105 observed with the Indian X-ray Astronomy Experiment (IXAE). The power density spectra of 130 independent data sets with 512 data points each, with a bin size of 100 msec were added to generate the plot at the top. A strong peak at a frequency of 0.7 Hz is clearly visible. In the two plots below that two power density spectra with peak at frequencies 0.62 Hz and 0.82 Hz respectively show the frequency variation in the QPO peak. The times of observations are given in the figure. The first harmonic at a frequency of twice that of the main peak is seen in all the three plots

Table 1. IXAE observations and results of GRS1915+105

Anticorrelation between the total intensity and hardness ratio (6-18 keV to 2-6 keV) is seen at very small time scale. Details of the analysis will be reported in a separate paper.

© European Southern Observatory (ESO) 1997

Online publication: June 30, 1998
helpdesk.link@springer.de